The Amsterdam Treaty: The Blueprint for the Future Institutional Balance?
The Amsterdam Treaty: The Blueprint for the Future Institutional Balance?
Abstract and Keywords
The 1996‐97 intergovernmental conference (IGC) aimed, inter alia, at solving the most pressing institutional issues in order to prepare the ground for the next wave of enlargement. The paper first contrasts this IGC's mandate with the outcome of the Amsterdam Treaty. It then analyses the four most significant reform steps with a view to democratic governance at the EU level: they concern the issue of ‘appropriate representation’ in the European Parliament; the appointment of the Commission President; the latter's powers concerning the internal organization of the Commission; and, finally, the new powers and competences of the EP. The authors conclude that the incremental institutional changes during the two decades since the first direct European elections amounted to a fundamental reform and that, in the future, this new inter‐institutional balance would be refined rather than fundamentally challenged.
The 1990–1 intergovernmental conference (IGC) was convened only twenty‐nine months after the last treaty reform had come into force.1 In the institutional field, which is our concern here, the 1991–2 IGC had by no means resolved all the pressing problems. In particular, some crucial questions relating to prospective further enlargements of the European Union (EU) had not been tackled.2 That those items which were explicitly postponed by the pre‐Maastricht IGC were but the ‘tip of the iceberg’ became obvious during the final phase of the negotiations leading to the 1995 enlargement, when the so‐called ‘Ioannina Compromise’ showed that the institutional structure had been stretched to its limits. In particular, the perspective of a further enlargement of the Union to the East and South made it clear that the institutional mechanisms (designed for a Community of only six members) which had so far only been adapted mechanically without changing the original principles, would not match up to the needs of a Union of some twenty‐eight member states. At various meetings held after the Maastricht Treaty was signed, in particular at its meetings in Cannes (June 1995), Madrid (December 1995), and Turin (March 1996), the European Council therefore put additional items on the agenda of the (then) forthcoming IGC.
In this chapter, we will first contrast the most important of those official tasks with the eventual Amsterdam Treaty. We will then outline in some detail the four most significant reforms in the institutional and procedural fields contained in the treaty: (p.16)
(1) ‘appropriate representation’ in the European Parliament (EP);
(2) the appointment of the Commission President;
(3) the latter's powers concerning the internal organization of the Commission; and
(4) the new powers and competences of the EP.
An Overall Assessment of the Amsterdam Treaty
Article 2, 5th indent of the Treaty on European Union (TEU)3 concerns ‘to what extent the policies and forms of cooperation introduced by this Treaty may need to be revised with the aim of ensuring the effectiveness of the mechanisms and the institutions of the Community’ (emphasis added). In that respect, the IGC did precisely what it had been asked to do. With regard to some of the new policies dating from 1991, the Amsterdam Treaty brings about procedural innovation (public health, vocational training, environmental policy, and trans‐European networks—TENs). No changes will take place concerning youth training, cultural, consumer, and cohesion policies. With regard to cooperation in the field of justice and home affairs (JHA), the IGC came to the conclusion that only a few issues were currently suitable for transfer to the first pillar (asylum, visas, and immigration). However, the new Title IV of the Treaty of the European Community (TEC) includes a series of deviations from the standard procedures of the first pillar; after five years, and particularly if a specific Council decision to this effect is taken, some of these peculiarities will disappear.
Another explicit task of the IGC, agreed in Maastricht, was to potentially widen the scope of the co‐decision procedure ‘on the basis of a report to be submitted to the Council by the Commission by 1996 at the latest’ (Article [189b] EC). In fact, the scope of the co‐decision procedure was significantly widened in Amsterdam, not least by the near abolition of the cooperation procedure. Furthermore, the procedure was simplified and improved, on the basis of experience gained since it had come into operation (see below). In this respect, too, the conference carried out its mandate.
The Brussels European Council (10–11 December 1993) had also asked the IGC to consider ‘ . . .any measures deemed necessary to facilitate the work of the institutions and guarantee their effective operation’. Only partial progress was made in this area (see below): the upgrading of the (p.17) directly elected EP through the reform and extension of the co‐decision procedure, the legal upgrading of the transparency rules in the treaty, and the new organizational structure for the common foreign and security policy (CFSP) may be seen as the main achievements of the 1996–7 IGC. On the other hand, the variety of available decision‐making procedures is still confusing, and there has been only a limited move from unanimity to majority voting.
Some of the IGC's tasks were clearly not fulfilled. For instance, the budgetary procedure is not mentioned in a single line of the new treaty, despite the fact that the Inter‐institutional Agreement of 29 October 1993 had requested the IGC to propose improvements to this procedure. When adopting the institutional provisions of the Accession Treaties admitting Austria, Finland, and Sweden, the member states and the applicant countries had furthermore agreed, in the communiqué of the December 1993 Brussels European Council that ‘the IGC to be convened in 1996 will consider the questions relating to the number of members of the Commission’. Although relevant also in the wider context of efficiency, the issue of the size of the Commission was postponed at Amsterdam.4 That the reweighting of the votes in the Council was also rescheduled must be seen in the context of the decreasing likelihood of the forthcoming enlargements happening as soon as initially envisaged by some. Given the clear differences between the smaller and larger states during the IGC, it was a pragmatic choice to postpone taking any decisions on those issues which will probably not prove to be of crucial importance before the next enlargement.
In general, such reforms as had been envisaged with a view to a further widening of the Union are still basically missing.5 The conference did not succeed in putting in place the necessary institutional changes which would allow the enlargement negotiations to concentrate on policy issues. Neither the distribution of the seats in the EP, nor the future size of the Commission, nor the future weighting of votes in the Council were tackled in a precise manner. Although Protocol 11 on the institutions indicates possible solutions, these remain rather vague and open to further discussion. In other words, the preparation for enlargement was postponed.
In short, looking at the institutional and procedural results of Amsterdam, the conclusion must be drawn that the IGC was both a success and a failure: it (p.18) successfully managed to agree on at least some important reforms, but it failed to prepare for the next enlargements.
The Most Significant Aspects of the Reforms
The 1997 treaty will not be remembered for particularly eye‐catching or, in the short term, far‐reaching institutional and procedural reforms. It does not represent a qualitative leap towards a more democratic system of EU governance. If the latter is understood as a regime in which the rulers are held accountable for their political actions by the citizens,6 the 1997 treaty continues along the path of incremental reform which, for a long time, has been the trademark of European integration. Once again, the EP is strengthened to some extent, and the electorate is thus somewhat better able to sanction the exercise of Euro‐politics in European elections (the frequently national character of EP elections remaining a crucial problem to be tackled). As regards transparency—another important aspect of democratic control—only marginal reforms were introduced. No improvements (or at least no direct ones) can be reported concerning further (possible) features of democratic governance, such as direct citizen participation7 or a collective citizens' identity.8
Although much clearly remains to be done to ‘democratize’ the EU,9 a number of remarkable innovations were agreed whose implications for the Union's institutional balance should not be underestimated. (Laffan 1997a: 302; Dinan 1997: 199)10.
The European Parliament and ‘Appropriate Representation’
As a former President of the EP has said, there is no other parliament like it (Dankert 1997: 212). One of its many special characteristics is that the citizens represented in it are not equal political subjects, in that the weight of a vote differs considerably between member states: whilst a German Member of the European Parliament (MEP) represents more than 800,000 citizens, a Belgian (p.19) MEP represents only about 400,000 people, and a Luxembourg MEP represents just 66,000 citizens.
From an institutional perspective, the EP's main goal has always been to become one of two ‘legislative chambers’ in the EU. Within this model, the EP would be the representative of the people of Europe, whereas the Council would represent the states. In its May 1995 report for the IGC, the EP made this perfectly clear when discussing the issue of reweighting votes in the Council: ‘ . . . it is in the Parliament that population is represented. Council represents States’ (EP Resolution 17 May 1995b, PE 190.441, pt. 22.iii).11 In other words, proportionality should not be the goal of the Council, but it should be improved in the EP. The Amsterdam Treaty deals explicitly with the issue of proportionality: Article 189  EC has been amended by a new second paragraph fixing the maximum number of MEPs at 700. Article 190 para. 2 [138a] EC, which now includes a table of the number of MEPs per member state, is complemented by a new subparagraph, as follows: ‘In the event of amendments to this paragraph, the number of representatives elected in each Member State must ensure appropriate representation of the peoples of the States brought together in the Community’ (emphasis added).
It should be noted that this rule was not intended to apply immediately, but only in the event of the amendment of Article 190 para. 2 EC (in other words, probably not before the next enlargement). The setting of a ceiling on the maximum number of MEPs, combined with the need to secure appropriate representation, ensured a tricky agenda for the IGC on enlargement: applying the current system, Poland's MEPs alone (representing some 38 million citizens) would occupy almost all of the remaining available seats. Since it is most likely that Poland will be among the next wave of EU members, and that it would not be the only new member, the enlargement negotiations, which began in March 1998 (Presidency conclusions of the Luxembourg Summit, 12–13 December 1997), will have to tackle the issue.12
While the stipulation on the maximum number of seats is clear and precise, the appropriate representation formula in Article 190 para. 2 EC is open to interpretation. Obviously, the negotiators of the founding Treaties thought that the terms agreed were appropriate, otherwise they would neither have signed the Treaties nor proposed that their respective parliaments should ratify them. The new provisions should not be regarded as simply superfluous, however. Rather, they may be interpreted as containing first, the acknowledgement of the basic idea that the representation of the population in the EP should also be appropriate in the future (i.e. when the rather rigid adaptation rule applied in all previous enlargement negotiations no longer applies, because (p.20) the maximum size of the EP has been reached);13 and second, it gives some sort of guidance for future negotiations on this issue, since the term may be analysed and interpreted. It is significant that Article 190 para. 2 EC does not read ‘equal representation’ or ‘proportionate representation’, but ‘appropriate representation’.14 This allows room for further (normative) considerations.
Two principles might be taken into account in order to achieve the goal of representativity. First, the vote of each EU citizen could be of approximately equal weight. However, this rule cannot be applied systematically if only 700 seats are available, because of the great differences in size between the member states: the ratio between the populations of Luxembourg and Germany is approximately 1:205. Given that there are other large member states in the Union, a total of 205 MEPs for Germany is out of the question, because the maximum of 700 MEPs would quickly be exceeded. Consequently, the number for Germany has to be lower than 205, which would mean that Luxembourg would not get a single MEP (and there are several other equally small states among the applicants, in particular Cyprus). Thus, the larger the Union gets, the more states would share the ‘Luxembourg destiny’ of not getting a single representative in the EP.
Second, since the building of new transnational electoral districts for the purpose of European elections is out of the question for the foreseeable future, the only alternative solution in order to attain at least ‘conditional proportionality’ is a minimum threshold rule: in order to ensure representation of all the (still separate15 and some quite small) peoples of the Union, at least one MEP would have to be attributed to each state. This rule, however, does not seem to ensure appropriate representation, since a single MEP cannot represent the political spectrum of his or her country of origin in accordance with the outcome of the (European) elections.16 At least three seats per country would (p.21) therefore be needed to ensure fair representation of a country's political forces.17
However, the application of these two principles (strict proportionality and a minimum of three MEPs per state) would mean that the five largest member states would gain from the changes, whereas all other countries would lose seats.18 In short, the proposed reform might be interpreted as corresponding to the EP's preferred option. However, Protocol 11 ‘on the institutions with a view to further enlargement’ annexed to the new treaty indicates that no corresponding reform (i.e. in the direction of ‘one country, one vote’) will be introduced in the Council in the foreseeable future; the system of weighted votes is still beyond question and will probably be further adapted towards greater proportionality. While improving proportionality in the EP in the manner foreseen by the Amsterdam Treaty will bring the EP closer to the principle of equal representation as outlined in democratic theory and as usually practised in national parliaments, it should be noted that only the largest countries (and Germany in particular) will be winners in practice. Reweighting the votes in the Council as indicated by the Treaty of Amsterdam would again be in the interests of the large countries only. Even when taking into account the ‘special case theorem’ (i.e. the EU as a sui generis system), it will be difficult to put the suggested reforms into practice, that is, to modify the traditional federal pattern of representing the population and the states in two separate chambers in a way which unilaterally strengthens the larger states. At least empirically, Joseph Weiler's (1988) assumption is still valid: so long as there is no redefined, integrated European polity, the social legitimacy of the European system remains precarious, and it is thus difficult to convince the members of the separate polities of Europe to give up power (in this case, seats in the EP) even for the sake of such widely acknowledged democratic principles as equal (or at least more ‘appropriate’) representation. Fundamental controversies over the institutional set‐up of the EU may be expected to come to the fore once the governments begin to tackle the details of the tricky issues that were postponed in Amsterdam.
As far as the Commission is concerned, two institutional innovations should be noted: a change in the procedure to nominate its President, and changes to the internal balance of the Commission.
The Maastricht Treaty reforms gave the EP the dual right to be consulted when the Commission President is chosen by the member governments, and to give its assent to the investiture of the college as a whole. Under the Amsterdam Treaty, the assent of the EP is needed for the appointment of the Commission President,19 and its approval is required for the new team of Commissioners. However, it is still the member governments who nominate the candidates. The nominee for President will be given a greater say in the choice of the members of the college. In the past, the nominee was only ‘consulted’ by the appointing member states, whereas the Amsterdam Treaty gives him or her a right of assent.20 This acknowledges the enhanced status of the President‐to‐be, who, by that point, will already have been the subject of a vote of confidence in the directly elected EP.
Given this new investiture procedure on the one hand and, on the other, the enhanced role of the President vis‐à‐vis his or her colleagues in the college of Commissioners (see below), we conclude that the character of the ‘appointing‐the‐Commission game’ might change in the medium term (see Hix, Chapter 6). Most likely, the presidential candidate will be invited to a hearing at the EP before the formal appointment takes place, and his or her political programme will in future be subject to parliamentary scrutiny. Due to this partial21 shift from secret diplomatic negotiations to a more open and public procedure, the terms of the debate on European policy‐making could be reshaped along party political lines, even though Delors' suggestion22 that each political party should nominate a candidate before the 1999 European elections was not followed in the short term. This might, in turn, prompt the development and public marketing of alternative ideological designs for the European agenda.
As already mentioned, the decision on the future size of the Commission was postponed at Amsterdam. In the eyes of many, however, the key question has less to do with the actual number of Commissioners and much more to do with the internal organization of the Commission. In this respect, several changes were agreed upon.
First, the role of the Commission President was strengthened. The new first sub‐paragraph of Article 219  EC reads: ‘The Commission shall work under the political guidance of its President’. It should be noted that ‘political guidance’ is less than a right to determine decisions or even to set compulsory (framework) guidelines (Bieber 1997: 240). Article 213 para. 2  EC on the independence of the Commissioners and Article 219 para. 2  EC on majority voting in the Commission remain unchanged. In the second sub‐paragraph of Declaration No. 32 attached to the Final Act ‘on the organization and functioning of the Commission’, the President's leading role is further qualified: ‘ . . . the President of the Commission must enjoy broad discretion in the allocation of tasks within the College, as well as in any reshuffling of those tasks during a Commission's term of office’ (emphasis added). This enhanced competence with respect to the allocation of tasks during the Commission's term in office, and not merely when the candidates are appointed, may lead to ‘a considerable power to discipline the otherwise undismissible members of the Commission’ (Wessels 1997: 127; our translation).
Quite apart from its probably undisputed effects with respect to the institution's efficiency and the coherence of its policies, both strengthening the Commission President by giving him or her organizational prerogatives and reorganizing the Commission's structure with a view to possibly centralizing specific policy areas in the hands of a few higher‐order Commissioners (see below) might lead to a more person‐oriented public perception of the Commission. Just as the heads and senior ministers of national governments are paid more attention than other members of a government, this might, in the medium term, result in more public awareness for some Commissioners, and thus for the entire Commission as a central political actor in the Union. In this context, it is noteworthy that the Commission's right of initiative is now expanded to the areas in the third pillar23 and the new Title IV on the free movement of persons—areas which are of particular interest to the EU's citizens as a whole.
In addition to dealing with the role of the Commission President, the IGC acknowledged and supported the Commission's intention to prepare a ‘reorganization of tasks within the College. . .in order to ensure an optimum division between conventional portfolios and specific tasks’ (Declaration No. 31 on (p.24) the Commission, sub‐para. 1). This reorganization was scheduled to be effective before the year 2000, that is, when the next Commission was due to come into office.
The reorganization will include a corresponding restructuring of the Commission's subdivisions (Declaration No. 31 on the Commission, sub‐para. 3). The only explicit target set by the IGC in this respect was that there should be a vice‐president responsible for all external affairs matters (in the 1995–9 Commission, responsibility for foreign policy and external commercial relations was divided between six Commissioners24). This must be seen in the context of the proposed new type of CFSP ‘troika’25 consisting of the Council Presidency, the new High Representative for the CFSP, and the external affairs Commissioner. According to Article 18 para. 4 [J.8] TEU and Article 27 [J.17] TEU, the Commission will be ‘fully associated’ with the work carried out in the CFSP field.26
The EP as Co‐Legislator
Reducing the EU's democratic deficit has often been equated with increasing the EP's legislative role, as is still the case even in recent writing, which also stresses other aspects of the EU's democratic deficit (see Grande 1997 for further references). In this respect, the Amsterdam Treaty introduces two changes: first, the reform of the co‐decision procedure and second, the upgrading of the involvement of the EP in many cases.
The reform of the co‐decision procedure (Article 251 [189b] EC) must be considered as one of the major achievements of the IGC. The following changes were agreed:
• If the Council accepts the EP's amendments at its first reading, it may adopt the act at this very early stage. Given the fact that carrying the procedure on to the (cumbersome and time‐consuming) conciliation procedure stage is rather unpopular among MEPs as well as ministers, this can be considered a major change in so far as the MEPs will be inclined to draft their amendments carefully, trying to anticipate the Council's views on the issue in order to get the approval of the governments already at this early stage.27
• If, in turn, the EP approves the Council's common position during its second reading, the act is deemed to be adopted without being referred back to the Council.
• If, in its second reading, the EP fails to take a position within three months, the act is deemed to be adopted in the version of the Council's common position; this seems to be the last remaining imbalance between the two ‘chambers’, since the decision finally taken would not, in this case, necessarily reflect a jointly agreed text.
• The EP can reject the common position directly during its second reading; that is, it no longer has to notify the Council in advance of its intention to reject. Consequently, the Council's ability to convene the Conciliation Committee at this stage was dropped.
• If, in its second reading, the Council accepts the EP's amendments, the act is deemed to be adopted without any further formal decision.
• A third reading still exists, but it was changed in such a way that the Council may no longer resume its original common position after a failure of the Conciliation Committee; this had been one of the main criticisms of the original co‐decision procedure.
• Various new time limits were introduced to ensure that the period between the EP's second reading and the outcome of the whole procedure does not take longer than nine and a half months; only the first reading will still be without time limits. In any case, experience shows that, due to the existing time limits, co‐decision is already the fastest procedure available, despite the possibility of three readings in both chambers.28
These changes show that, on the issue of the reform of the co‐decision procedure, the EP came very close to obtaining what it had asked for in the IGC. In fact, the changes to a very large extent eliminate the procedural imbalances between the two major players, the Council and Parliament. Any remaining differences can be perceived as a functionally useful distribution of roles between two legislative chambers, while the overall political weight of the two institutions within the co‐decision procedure may now be considered equal. Moreover, it can be argued that the last significant formal inequality—the requirement for the EP to make a pronouncement on the Council's common position within a reasonable period—mainly represents a strong incentive for effective internal organization. Furthermore, it seems plausible that cases of parliamentary non‐decision in the second reading will, in fact, indicate an overall positive rather than a hostile reaction by the EP to the common position: those MEPs who are definitely against the measure as shaped by the Council are more likely to ask for far‐reaching amendments than to block EP decision‐taking, since a blockade leads directly to the adoption of the unwanted (p.26) act. The only realistic scenario in which no decision might be taken by the EP would seem to be when both immediate approval and also the making of amendments enjoy considerable support among MEPs. In such a case, it is possible that neither the absolute majority of votes cast (for approval) nor the absolute majority of MEPs (for the amendments) might be reached, because those who want to approve know that their preferred outcome is the default solution, and they may therefore not be ready to compromise. In addition, it is quite likely that there will be a majority for at least some (minor) amendments and thus a decision. The EP's performance in recent years shows that it has almost always been successful in producing opinions on time. The pressure on the EP is obviously great, since its very reputation is at stake; the MEPs can hardly press for greater participation in the legislative process if they fail to fulfil their allotted tasks.
It should be noted, however, that the new wording of Article 251 [189b] EC does not contain a parallel provision to cover the Council not acting within the time limit of three months in its second reading: if the Council fails to reach a decision on the EP's amendments in time, the procedure simply comes to a halt,29 but the legislative act is not deemed to be adopted according to the EP's version as it is in the opposite case. At this stage of the legislative procedure, however, the Commission's opinion has an important role to play, and it would therefore not be clear which version of the text was the latest: the EP's or the version amended by the Commission. From this perspective, strict equality between the two legislative chambers would not offer an adequate solution to the peculiarity of the triangular institutional set‐up.
Against this background, we conclude that the reform of the co‐decision procedure finally puts the EP on an essentially equal footing with the Council (at least as far as this procedure is concerned; see also Wessels 1997: 128). Given that this procedure will be applied in many more cases than before (see below), the 1996–7 IGC is indeed a major step towards a bicameral legislative model (with two strong players) at the EC level. It is interesting to note that the wording of the treaty did not take this gradual shift from a strongly Council‐centred to a more balanced bicameral system into account: while in the Single European Act (SEA) the formula ‘the Council shall. . .in cooperation with the European Parliament adopt . . . ’ was used, the Maastricht Treaty does not mention the EP in most cases: ‘The Council shall, acting in accordance with the procedure referred to in Article 189b. . . , adopt. . . ’.30 The Amsterdam Treaty will not change the remaining imbalance in the wording of the treaty, (p.27) although the titles of the legislative acts based on Article 251 EC already include the EP (e.g. ‘Directive. . .of the European Parliament and the Council of Ministers on . . . ’).
Having started out as a purely consultative body composed of representatives delegated from each of the national parliaments, the past few decades have seen an extension of the EP's competences with each major treaty reform. In the aftermath of the first direct elections in 1979, the EP itself gave an important impetus to the 1986 SEA which introduced both the cooperation and the assent procedures, although its ‘Draft Treaty on European Union’31 was not accepted as such by the governments. Nevertheless, pure consultation was retained as an alternative mode of EP involvement for many areas of European policy‐making. The Maastricht Treaty continued along these lines by adding yet another procedure; ‘co‐decision’ under Article 251 [189b] EC gave more far‐reaching (although not yet equal ) powers to the EC's parliamentary chamber, but it was far from representing the one and only standard procedure. The result of this incremental reform process was a patchwork‐style landscape of EC decision‐making: in each case, one of approximately twenty variants32 of the four main procedures applied. None of them, however, put the EP on an equal footing with the governments represented in the Council. Therefore, the EP's demands for the 1996–7 IGC focused not only on improving co‐decision, but also on making it the only standard procedure in all EC law‐making.
Yet again, no uniform decision‐making procedure was established by the Amsterdam Treaty. However, many specific changes were introduced, among which we would like to highlight the following points.
In the TEU, Article 7 [F.1] gives the EP an additional right of assent when the Council determines a breach of the Union's principles by a member state. By contrast, there is still almost no involvement, or at least only consultation, in the second and third pillars of the Union: with regard to some areas of the third pillar (Article 39 para. 1 [K.11] in connection with Article 34 para. 2 [K.6] TEU), a special consultation procedure will apply, which already exists under one provision in the first pillar, namely in connection with the conclusion of international agreements (Article 300 para. 3  EC). The Council may lay down a time limit (which shall be no less than three months) for the delivery of the EP's opinion; in contrast to the ordinary consultation procedure, the Council may act without this opinion if it is not delivered in time (see Article 39 para. 1 TEU). In those areas which will be (p.28) ‘communitarized’, that is, transferred from the third to the first pillar,33 the EP will not be involved at all, or will at most be involved on a consultative basis during the five years after the entry into force of the new treaty. After five years, the competence to harmonize the rules and procedures concerning visas will automatically be subject to the co‐decision procedure, whereas for the rest of the decisions taken under the new Title on free movement of persons, this transfer has to be decided unanimously by the Council after consulting the EP (see Article 67 para. 2 EC).
In those areas already covered by the EC, the reform did not bring about a single legislative procedure. The state of affairs after Amsterdam can be summarized by the following five points. First, with regard to certain central issues where the Amsterdam Treaty introduces new provisions or amends existing ones, the EP is still only consulted (e.g. authorizing the establishment of closer cooperation among some member states) (Article 11 para. 2 EC), measures outlawing discrimination (Article 13 EC), most decisions on asylum and immigration (Articles 64 and 67 of the new Title on the free movement of persons), and parts of the social and research and development (R&D) policies.
As far as those articles which remained unchanged by the Amsterdam Treaty are concerned, three provisions should be pointed out:
1. The new main legal basis for action in the field of agricultural policy (Article 37  EC), will also provide for consultation only in the future. However the provision on public health (Article 152 para. 4  EC) was changed to include areas previously governed by Article  EC, that is, veterinary medicine and phytosanitary measures. Since decisions in the area of public health are already being taken under the co‐decision procedure, we may conclude that the EP now has at least ‘a foot in the door’ on the way to becoming a decisive co‐legislator in the agricultural field. However, in the important area of the common agricultural market organizations, the Council acts after having merely consulted the Parliament.34 This is particularly deplorable since, in this area, neither the EP nor the national parliaments35 have a say with regard to the content or the budget of the policy.
2. Another important area in which the EP is still involved on a consultative basis only is the harmonization of legislation concerning indirect taxes, such as turnover taxes and excise duties (Article 93  EC).
3. Finally, Article 308  EC remains unchanged. It allows the Council to adopt legislation which cannot be based on more specific competences but which is nevertheless considered ‘necessary to attain, in the course of the operation of the common market, [if] one of the objectives of the Community and this Treaty has not provided the necessary powers’. This is the so‐called ‘subsidiary competence’ provision, which has provided the legal basis for many important legislative acts.36
Second, the cooperation procedure has been replaced by co‐decision in most cases, but not in relation to Economic and Monetary Union (EMU) (Articles 99 para. 5; 102 para. 2; 103 para. 2; and 106 para. 2 EC). It should be remembered here that ‘cooperation’ gives the EP no power of veto.
Third, although co‐decision applies to only eight of the 36 new competences attributed to the EC/EU by the Amsterdam Treaty, there are 14 (15)37 issues where the procedure has been changed to co‐decision. These are in addition to the 15 set out in the Maastricht Treaty (see the list in Annex V.b of the 1995 Council report, footnote33). In future, the EP will thus be a co‐legislator under 37 (38) provisions altogether; included among these are central legislative powers of the Union, such as most common market‐related provisions, and at least some types of decision in most of the other policy areas (with some exceptions and qualifications however, as outlined in the two preceding and subsequent points).
Fourth, there are also some new cases of non‐involvement of the EP, however: for example, when decisions are taken on further member states joining an existing form of ‘closer cooperation’ (Article 11 para. 3 EC), on emergency measures relating to immigration (Article 64 EC in the new Title IV on the free movement of persons), on recommendations on employment policy (Article 128 para. 4 EC in the new Title IV on the free movement of persons), on the implementation of social partner agreements in social affairs (Article 139 [118b] para. 2 EC), on adapting or supplementing R&D programmes (Article 166 para. 2 [130i] EC), and on the suspension of rights deriving from the EC Treaty for a member state which is in breach of fundamental principles of the Union (Article 309 EC). In the field of economic policy too, the EP is (p.30) often only informed of decisions taken by the Council (e.g. Articles 99, 100, 104, and 114 EC).
Fifth, it is worth noting that no changes were agreed on the EP's participation in the budgetary procedure, although this issue had been put explicitly on the IGC agenda by the Interinstitutional Agreement of 1993 (OJ 93/C 331/1 of 7 December 1993; see also EP Resolution PE 190.441, pt. 34 of 17 May 1995b). Furthermore, the EP will still be involved in CAP reform and tax harmonization on a consultative basis only (see above). Even under the Amsterdam Treaty, therefore, the EP is still far from being a co‐equal player in financial matters. On the other hand, the EP's involvement in budgetary matters has been extended to some degree. Third pillar operational expenditure will now fall under the EU budget, unless the Council decides otherwise (Article 41 paras. 2 and 3 [K.13] TEU); the same applies to second pillar operations except military actions (Article 28 paras. 2 and 3 [J.18] TEU); and finally, the EP will have more influence under co‐decision with respect to the fight against fraud (Article 280 [209a] EC).
In attempting an overall assessment of the extension of EP competences under the Amsterdam Treaty, it should be remembered that few had expected the IGC to result in a ‘landslide’ regarding EP competences in EC/EU decision‐making. Nevertheless, significant improvements from the EP's point of view were agreed. We consider it likely that co‐decision will henceforth be perceived as the future standard procedure, with consultation or cooperation soon being considered as the exception to the rule. In future, this might not only change the public perception of the EP's powers vis‐à‐vis the Council, but might also make it easier to switch to co‐decision as the single legislative procedure at a later IGC.
For the moment, however, under the Amsterdam Treaty, the list of areas where the Council is authorized to act without being dependent on the EP's approval is still considerably longer than the co‐decision list (Duff 1997: 146ff.). Of course, a proper assessment should not only consider numbers, but must also take into account the importance of the respective competences. In this respect, it should be noted that a large proportion of the ‘traditional’ competences of the old European Economic Community, in particular those related to the establishment of the common market (the ‘four freedoms’), have been shifted to co‐decision, including the prominent Article 95 [100a] EC. Turning to the newer competences, the picture is quite different. Like many national parliaments, the EP is involved to a lesser extent in foreign policy (international agreements and the CFSP in general), monetary policy
(e.g. defining the statute of the European Central Bank—ECB), classic governmental activities (such as emergency measures and amending programmes during their execution), and where corporatist patterns prevail (in social policy). However, it is easy to overlook the fact that there is also a long (p.31) list of matters in which the Council acts as a genuine legislator without Parliament acting as a co‐legislator: for example, the EP is merely consulted in the field of agricultural policy; in the area of the third pillar, whether ‘communitarized’ or not; with regard to some decisions in the field of social policy and also tax law; and even in the classic common market‐related areas of competition law, state aids and transport policy. One of the major wishes of the EP, namely that the co‐decision procedure (Article 251 [189b] EC) should apply when the Community institutions act as legislators (see EP Resolution PE 190.441, pt. 29 of 17 May 1995b), was thus only fulfilled to a quite limited extent.
In even sharper contrast to national parliaments, however, the MEPs have only a limited say in quasi‐constitutional matters, such as decisions regarding flexible cooperation and the suspension of rights deriving from the treaties. Most strikingly, the EP is not involved at all in amendments to the EC/ EU's ‘constitution’ (i.e. treaty reform conferences).38 On the one hand, the two chamber set‐up of the EC/EU (i.e. the collaboration of the Council and the EP in the legislative process—the Commission's role is a completely extraordinary one) resembles national systems in many instances, while on the other hand, some very specific features still prevail. Might there be a coherent hypothesis to explain this dichotomy? First, it seems plausible to argue that each IGC brings the EC/EU decision‐making structure a little closer to a federal state model, with a parliamentary and a state chamber. However, this statement needs to be qualified by saying that the European case is a unique type of federal decision‐making structure, at least so far. Furthermore, this step‐by‐step approach always lags behind the parallel increase of competences of the Union. In most cases where a new competence is introduced, the default solution is still that the EP is outside the decision‐making core, and that the Council decides alone. With a few exceptions,39 it seems to take at least one40 if not more41 IGCs in order to get the EP involved on an equal footing (p.32) with the Council. Again, the Amsterdam Treaty followed this trend, for example with respect to the (partial) communitarization of the third pillar, most ‘competences’ in the new area of employment policy, and many other changes. This suggests that the EP might only be involved to a greater degree after the bulk of decisions in the area have already been taken. Most common market‐related issues have already been decided; what remains is (to a certain extent) ‘peanuts’ in comparison to what was decided pre‐1992. By contrast, current issues, such as those falling under the third pillar, will be settled over the next few years, and the EP might again only be involved once the new policy has by and large been implemented. The same might be true for EMU, employment policy and tax harmonization.
To sum up: in the political context of the IGC, most commentators agree that the EP's role was upgraded by the new treaty, and many see it as the winner in the Amsterdam process (Dehousse 1998; Brok 1997: 211; Nickel 1997: 220; Wessels 1997: 130; Schönfelder and Silberberg 1997: 209). Viewing the changes in the light of the many remaining imbalances which still favour the Council, however, it is impossible to come to an enthusiastic overall assessment of the Amsterdam changes, when these are compared to the EP's original demands: in most areas of its activities, the Union is still a long way from being a balanced bi‐cameral legislature.
Another key issue in the pre‐IGC discussions was the plea for a reduction in the number of different decision‐making procedures. The most radical call came from the EP, which asked for an almost uniform decision‐making procedure in legislative matters, reducing the number of procedures to three: co‐decision as the norm, assent for constitutional matters and consultation for the CFSP (European Parliament, Resolution PE 190.441, no. 29, 17 May 1995b).
The Amsterdam Treaty did not meet these expectations. Whilst replacing the cooperation procedure (Article 252 [189c] EC) with co‐decision (Article 251 [189b] EC) in most cases, it left the provisions on EMU untouched. Furthermore, the new Treaty added some new variants, such as the new procedure for adopting the statute of the MEPs,42 or a (limited) right of initiative of the member states, which has now been introduced into the first pillar (Article 67 para. 1 EC). Moreover, the consultation and assent procedures, as well as many other variants, still persist.
It is easy to overlook the overall significance of the institutional arrangements decided at Amsterdam, because they appear rather unspectacular when viewed separately. Viewed from a distance, however, the Amsterdam Treaty constituted a major step forward in a process begun by the SEA and continued by the Maastricht Treaty: that of making the EP a co‐legislator with powers equal to those of the Council. Until the Amsterdam Treaty, co‐decision was biased in favour of the Council, and it was the exception rather than the rule in EC policy‐making. Our analysis has shown that the new co‐decision procedure puts both ‘legislative chambers’ on an equal footing. Due to the continued coexistence of other decision‐making procedures (not only in the second and third pillars, but even in the first), in many areas of activity the Union is still far from being a two‐chamber system, as demanded by the EP. But although there is still a long way to go quantitatively, it is plausible to argue that, in a qualitative sense, co‐decision will in future be perceived as the paradigm of EC decision‐making.
In that sense, the incremental changes to the EC's institutional balance from 1979 (the date of the first direct elections to the EP) to 1999 (the date of the entry into force of the Amsterdam Treaty) amounted as a whole to a fundamental reform whose basic principles are by now clearly outlined. This does not imply that incremental reform will no longer take place. On the contrary; further adaptations, such as the extension of co‐decision to those legislative areas not yet covered, seem to be useful (with a view to improving democratic accountability, at least to some extent43) and likely further steps.
Starting from the neo‐institutionalist assumption44 that, once in place, institutions (in the wider sense) shape not only the strategies but even the preferences of political actors, and thus usually represent trajectories for further developments, the new pattern of institutional balance under co‐decision as established by the Amsterdam Treaty may be expected to basically remain the same: first, a much more politicized Commission, drawing legitimacy from the way in which it came into office; second, an EP which basically plays an amending role, but nevertheless has a decisive say regarding the life and death of legislative acts; and third, a Council of Ministers which is no longer in a position to de facto impose its views on the other institutions in EC decision‐making, most notably the EP.
(p.34) There are as yet no signs that, even in the medium term, a new compromise formula between the EU governments might lead to a fundamentally different relationship between the main institutions. This statement may be substantiated by the following considerations. First, it seems unlikely that the EP will acquire predominance over the Council in the foreseeable future. No relevant political actor (not even the EP) actively promotes a model such as the Austrian system of a directly elected Nationalrat (Parliament) and a second chamber, the Bundesrat (representing the Länder) with a veto right restricted to only a few cases. Bearing in mind that all further treaty changes will have to be negotiated by the governments sitting in the EU Council, politically it is naïve to expect anything beyond an extension to further areas of EU policy‐making of the current equal position of the EP and the Council under co‐decision.
Second, the EP has dropped its demand to gain a genuine right of initiative, which would have ended the Commission's monopoly in this area.45 Evidently, an EP right of initiative would alter the institutional balance considerably, since the Commission would be almost ‘out of the game’ whenever the EP took the initiative. However, in order to make it more than just a formal right, the EP would have to change its working structure fundamentally, with a view to creating detailed draft legislation. At the moment, no such political will exists, not even among the MEPs; nor is there sufficient working capacity. The present division of labour between an agenda‐setting Commission and two decision‐making bodies seems to be widely accepted, and thus unlikely to be changed in the near future.
Third, the problem of how to reweight the votes in the Council is still on the agenda. Whatever the outcome of future negotiations, it seems plausible to expect that it will not fundamentally change the Council's position within the institutional triangle, nor will it alter the character of that institution to such an extent that it affects the Union's decision‐making system as a whole. The debate has two opposite poles, demanding that the Council be made either a fully representative body, or else a US Senate‐like body with equal weight for all states. It is probable that neither of these extremes will be put in place, since the former is strictly opposed by the smaller member states, and the latter by the larger ones. Therefore, only minor changes in the middle ground are to be expected, without fundamental consequences for the status and functioning of the Council.
These arguments underpin our hypothesis that the new treaty (particularly its reform of the co‐decision procedure) outlines an interinstitutional balance (p.35) for the EU's legislative process which is likely to be refined rather than fundamentally challenged in the foreseeable future.
(1) This could have come as a surprise only to the uninitiated, as Art. N para. 2 of the Treaty on European Union (TEU) called for a conference of representatives of the governments of the member states to be convened in 1996, with a view to examining those provisions of the Maastricht Treaty for which revision was envisaged from the outset. In addition to the effectiveness of the EC's mechanisms and institutions, the revision was thus planned to review notably the Common Foreign and Security Policy (the second pillar) and Justice and Home Affairs (the third pillar). See Arts. [B 5th indent; J.4 para. 6 and J.10 TEU and Art. 189b para. 8 EC] and Common Declarations No. 1 and 16 annexed to the Final Act of the Maastricht Treaty.
(3) In this text, we refer to the new numbers of all EU and EC Articles as agreed in Amsterdam. The numbers of Arts. in force before this reform which no longer exist are included in square brackets.
(4) Art. 1 of Protocol No. 11 ‘on the institutions with the prospect of enlargement of the European Union’ annexed to the Treaty of Amsterdam envisages that, at the date of entry into force of the first enlargement of the Union, the Commission will comprise only one national per member state. This new distribution of seats in the Commission is, however, conditional: the five larger states will give up their second Commissioner only if the issue of the weighting of votes in the Council of Ministers has been settled by the same date.
(5) The European Council, meeting in Corfu on 24–25 June 1994, had asked the IGC to consider ‘ . . .any other measure deemed necessary to facilitate the work of the institutions and guarantee their effective operation in the perspective of enlargement’.
(7) For specific recent contributions on this topic, see notably Abromeit (1998) and Nentwich (1998). Grande (1997) and Zürn (1996) also discuss these aspects in addition to others related to democracy at European level.
(10) Due to limitations of space, we can only highlight a few aspects of the Amsterdam reforms. For a much more detailed account, including many tables and detailed references, see Griller et al. (1999).
(12) As regards the legal value of this clause, the new wording does not bind future negotiators when trying to find a compromise formula on which all will be able to agree.
(13) The rule may be summarized as follows: A new member state should be given a number of MEPs which is smaller than or equal to the number of the next largest member state (in terms of size of population) and larger than or equal to the number of the next smallest member state; the details are open to negotiation. This rule obviously leads to a constant increase in the total number of MEPs.
(14) Bieber (1997: 239) argues in this respect that the term ‘appropriate’ is so vague as to be unhelpful in deciding the question of whether it means ‘proportionate’ or ‘minimum’ representation. He therefore concludes that the Amsterdam Treaty in fact adds nothing to the present situation. As we have argued, we look on the new wording as a point of departure for further consideration and negotiations.
(15) The term ‘separate’ refers to the widely shared perception that there is (as yet) no single ‘people’ of Europe, but rather a series of peoples of the different European states.
(16) As Duff (1997: 150) puts it: ‘While a reduction for all is inevitable, the importance of the small states having enough MEPs to allow for a fair representation of the major strands of political opinion must not be overlooked’. For a different system to ‘reconcile the principles of “one man one vote” with the need to proect small state representation’ (Laffan 1997a: 303), see the EP's De Gucht Report, summarized in Laffan (1997a).
(17) The three MEPs would not necessarily come from the centre, the right, and the left political movements of a country; if one particular party were very strong, it might send all three MEPs to Strasbourg. In our view, this would still be an ‘appropriate’ representation of this country's political spectrum, even though a minority might not be represented. Given that three representatives for a state is clearly a lower limit, the rule proposed in our text is intended to avoid the extreme case where, if only one seat were available, three almost equally large parties in a country would be ‘represented’ by the party which is the largest of the three (if only by a few per cent of the vote).
(18) Most striking would be Germany's enormous gain in seats: an additional 32 or 38 (depending on the method of calculation, and taking account of a minimum number of seats for the smallest member states).
(19) Art. 214 para. 2 sub‐para. 1 EC: ‘The nomination shall be approved by the European Parliament.’
(20) Art. 214 para. 2 sub‐para. 2 EC: ‘The governments of the Member States shall, by common accord with the nominee for President, nominate . . . ’. In practice, one can imagine the nominee being present at the decisive negotiations in the Council (meeting as the representatives of the governments of the member states) and giving his or her final and formal approval to the decision.
(21) The shift is partial, in so far as it is still the member states which, by common accord, choose the nominee for Commission President; in the earlier phases of the investitute procedure, the EP is thus involved at best indirectly, as the member states will anticipate the EP's stance to some extent.
(22) See e.g. Der Standard, 26 June 1998.
(23) It should be noted that, in contrast to the policy areas dealt with in the first pillar, the Commission shares the right of initiative with the member states in the third pillar.
(24) Responsibility for foreign policy and external commercial relations is divided up as follows between the Commissioners: Santer—CFSP; van den Broek—CFSP; Pinheiro—Lomé, ACP; Marin—Mediterranean states, Near and Middle East, Latin America, parts of Asia; Brittain—Commonwealth, foreign commercial relations; and Bonino—humanitarian aid.
(25) Note that the old troika between the former, the current, and the next presidency has been replaced by a new leading group which might be called a ‘duika’ (a two‐headed horse), consisting of the current and the next presidency only.
(26) This means that the Commission will be informed of, and consulted about, all activities in the CFSP field, but will have no right of initiative, as it does in the first pillar.
(27) We are grateful to Michael Shackleton of the EP for this argument. On the other hand, there are also good reasons for doubting that the EP will indeed react like a rational actor (we are grateful to Karlheinz Neunreither for this point).
(28) Information given by Andrea Pierucci, European Commission.
(29) Such an outcome might eventually be challenged before the European Court of Justice (ECJ), either via the Art. 230  EC procedure (by challenging the final result for procedural shortcomings), or via Art. 232  EC (failure to act, in infringement of the treaty).
(30) There are a few exceptions to this rule (see Arts. 229, 230, 241, 251 para.1 and 253 EC). All of these cases, however, are general provisions, and contain no specific power to act.
(31) This draft for a Treaty on European Union with innovative decision‐making procedures and institutional provisions, adopted on 14 February 1984 (OJ 84/C 77/33), is also known as the Spinelli Draft Treaty, because it was initiated by the so‐called Crocodile Group, led by the Italian MEP, Altiero Spinelli.
(32) The exact number depends on the way in which one counts; the subcategories mainly account for variations in the majority applicable in the Council and the various bodies consulted.
(33) See the new Title IV EC on visas, asylum, immigration and other policies related to free movement of persons, which includes the Community's competence, for example, to take limited emergency measures against immigration, or to decide on uniform visa rules; see Arts. 64 and 67 of the new Title.
(34) The EP's position during the IGC as formulated by its two representatives, Elisabeth Guigou and Elmar Brok, seems to have been that decisions of principle at least, but not necessarily decisions on the day‐to‐day business of agricultural legislation, should be taken according to the co‐decision procedure.
(35) This is because the Community has an exclusive competence for the agricultural policy, which is financed entirely by the EC budget. Furthermore, Council decisions in this field are taken by qualified majority, which means that national parliaments wishing to influence the decisions taken at Community level by binding their Minister for Agriculture may be unsuccessful, because he or she can be outvoted in the Council.
(36) In Annex VIII to its Report on the Operation of the Treaty on European Union of 6 April 1995, the Council counted 86 cases in the period from 1992 to spring 1995 where Art. 308 EC had been invoked, in areas such as the establishment of a number of bodies, in financial issues and in other areas as varied as the consolidation of existing Community legislation, the Community trade mark, and budgetary discipline.
(37) If one includes Art. 67 para. 3 of the new Title IV EC on the free movement of persons, which will apply after five years, another case must be added; hence the numbers 15 and later 38 in brackets.
(38) It should be mentioned, however, that the EP has to give its assent to the admission of a new member, and therefore indirectly to those treaty changes which are necessary in order to incorporate the new member into the treaty framework (see Art. 49 [O] TEU).
(39) For example, cultural, consumer protection and public health policies, as well as the Trans‐European Networks (TENs) were introduced by the Maastricht Treaty, and immediately made subject to the co‐decision procedure. It is, however, arguable that these cases are not important exceptions: the competences are rather limited, and were already partly covered by other provisions, as was the case with consumer protection.
(40) For example, development cooperation was introduced by the Maastricht Treaty subject to the cooperation procedure, and upgraded to the co‐decision procedure by the Amsterdam Treaty. Furthermore, environmental policy was introduced by the Single European Act subject to the consultation procedure, and upgraded by the Maastricht Treaty to co‐decision.
(41) For example, the provisions on the European Social Fund were inserted into the Single Act subject to the cooperation procedure, and only upgraded to co‐decision by the Amsterdam Treaty. This might also be the case with monetary policy and industrial policy, both of which were introduced by the Maastricht Treaty, but left untouched by the Amsterdam Treaty.
(42) This procedure entails an Opinion from the Commission, the approval of the Council acting unanimously, and decision by the EP.
(43) Without doubt, however, a democratic, political EU system in the wide sense of the recent debate on the ‘democratic deficit’ would demand many more changes. For a number of innovative proposals on how to democratize the Union, see e.g. Zürn (1996), Grande (1997), Fo/llesdal and Koslowski (1997), Weale and Nentwich (1998).
(45) Compare the EP's report of 17 May 1995 on the functioning of the Union with a view to the 1996 IGC (PE 190.441, pt. 21.I) with the Herman Report of 9 February 1994 (PE 203.601/endg. 2, Art. 32 of proposed Constitution).